![]() |
|||
|
|||
| :
News : Reviews : Editorials : About : Contact : Advertising : Privacy : Links : Forums
|
|
Q&A with Intel's George Alfs |
|
| An IDF Kickoff | |||
|
by Josh Walrath |
|||
|
Will Intel ever join the HyperTransport Consortium? To most people, it looks like PCI-E and HyperTransport are more complimentary than competitive. Even though this consortium was started by AMD, wouldn’t Intel find some very practical uses in its networking and chipsets divisions for the HyperTransport technology? Alfs: PCI Express is the latest technology that extends the bandwidth and performance of I/O to the next generation of computing and communication platforms. It is backed by the PCI-SIG with more than 850 industry-leading companies as an open industry standard. Intel supports such an industry-wide standard, and has launched our newest generation of 925/915 desktop chipsets supporting this standard in June'04. All PC systems shipping today with these Intel chipsets are supporting this latest I/O technology. A number of components such as add-in graphics cards and Gigabit Ethernet chips are already available today supporting PCI Express, and an increasing number of them are coming to the market everyday. HyperTransport technology is supported by a much smaller group of companies for their specific product needs. Due to our use of PCI Express, we don't intend to use it and I know of no plans to join that consortium. While Intel may have had some interesting bumps in the road concerning memory technologies, its embrace of DDR technology has enabled chipsets like the i865 and i875 to dominate the market. DDR-II now looks to be the next dominant memory type. What is Intel looking at down the road past DDR-II? Does the Yellowstone technology from Rambus hold any kind of interest? Alfs: Right now we are looking at moving up the speeds and capabilities of DDRII. We will be discussing that at IDF this week. After that, we are in discussions with memory vendors but we have nothing to announce right now. Does Intel see any advantages to its own implementation of X86-64 (EMT64 if I am not mistaken)? As technology moves forward, do you foresee Intel expanding the instruction set and abilities of IA-32e as Intel has expanded such things as SSE (now up to SSE-3)? Alfs: EMT64 is one of many features we bring to our platforms, along with Hyper-Threading Technology, DDRII memory, PCI Express, 7.1 audio, etc. etc. At this point, we are working with Microsoft to minimize compatibility issues out in the software community before they go gold. That is priority 1, 2, and 3 at the moment. How well supported is the Azalia specification? Was there a lot of communication between the different chipset and sound providers, or was this a wholly Intel funded project? When will we see the Azalia specifications released onto the mobile platforms from Intel? Alfs: Intel has worked with the industry to develop the High Definition (HD) Audio specification (formerly codenamed “Azalia”) for integrated audio that is capable of delivering the features and high-end performance of an add-in audio card. More than 80 participating companies, including PC and CE manufacturers, codec vendors, software providers and other industry leaders have teamed with Intel to develop the HD Audio v.1.0 specification—which any codec vendor can implement on a system supporting HD Audio. Intel HD Audio delivers significant improvements over previous generation integrated audio and sound cards, it is capable of delivering the support and sound quality for up to eight channels at 192 kHz/32-bit quality, while the AC‘97 specification can only support six channels at 48 kHz/20-bit. We are seeing lots of interesting HD Audio solutions for desktop PCs appearing in the market and are working to implement this technology into mobile form factors. This week should be a very interesting one for Intel, and for enthusiasts looking for high performance computing solutions. While some out there are viewing Intel as a “me too” company at the moment in regards to dual core technology, I think it is pretty obvious that Intel has had this concept well thought out before any announcements from AMD. Intel also provides more than 80% of x86 processors to the market, and the PC market today would not be anywhere without Intel. Even if AMD decided to produce all Sempron processors that are 86 mm square at their Fab 30, they can only approach supporting around 30% of the market. When analysts speak of the 800 pound gorilla, they are not kidding. While I may not agree with everything George says, he does have the advantage over me by actually working at Intel and seeing roadmaps and memos that I obviously can’t get my hands on. I do agree that Centrino is a great product though, and I am excited to hear about what will be coming up soon for that architecture. With AMD getting stronger in the mobile arena with the Athlon 64 Mobile series of processors, I think it is a sure bet that Centrino will be getting EMT64 within the year to keep their product competitive in time for Windows XP 64’s release in early 2005. I would like to thank George for allowing me to interview him with several lengthy questions. It was also great that he was able to get these questions back to me before the beginning of this Fall’s IDF. It has been a pleasure to work with him, and I hope to do so again shortly.
If you have found this article interesting or a great help, please donate to this site.
Copyright 1999-2004 PenStar Systems, LLC. |
|||